CLEAR: Revealing How Noise and Ambiguity Degrade Reliability in LLMs for Medicine

TL;DR

This paper introduces the CLEAR framework to evaluate how noise and ambiguity impact the reasoning and reliability of medical LLMs, revealing significant limitations in current benchmarks.

Contribution

The paper presents a novel evaluation framework that systematically assesses the effects of ambiguity and uncertainty on medical LLMs' performance, highlighting key limitations of existing benchmarks.

Findings

Increasing plausible answers reduces model accuracy and abstention ability.

Framing of abstention options affects model caution, with IDK increasing errors.

Scaling models does not fully address reliability issues, indicating a humility deficit.

Abstract

Medical large language model (LLM) evaluations rely on simplified, exam-style benchmarks that rarely reflect the ambiguity of real-world medical inquiries. We introduce the CLinical Evaluation of Ambiguity and Reliability (CLEAR) framework, which assesses how decision-space presentation, ambiguity, and uncertainty affect LLMs' reasoning on medical benchmarks. CLEAR systematically perturbs (1) the number of plausible answer options, (2) the presence of a ground truth or abstention option, and (3) the semantic framing of answer options. Applying CLEAR on three benchmarks evaluated across 17 LLMs reveals three notable limitations of existing evaluation methods. First, increasing the number of plausible answers degrades a model's ability to identify the correct answer and abstain against incorrect ones. Second, this lack of caution intensifies as the framing of abstention shifts from assertive rejection like "None of the Above" to uncertainty admission like "I don't know" (IDK). Notably, just including IDK in the answer space increases incorrect answer selections. Lastly, we formalize the performance gap between identifying the correct answer and abstaining from incorrect ones as the humility deficit, which worsens with model scale. Our findings reveal limitations in standard medical benchmarks and underscore that scaling alone does not resolve LLM reliability issues.